Third generation partnership project (3GPP) mobile communication systems based on a wideband code division multiple access (WCDMA) radio access technology are widely spread all over the world. High-speed downlink packet access (HSDPA) that can be defined as a first evolutionary stage of WCDMA provides 3GPP with radio access technique that is highly competitive in the mid-term future. However, since requirements and expectations of users and service providers are continuously increased and developments of competing radio access techniques are continuously in progress, new technical evolutions in 3GPP are required to secure competitiveness in the future. Reduction of cost per bit, increase of service availability, flexible use of frequency bands, simple structure and open interface, proper power consumption of a user equipment, and the like are defined as requirements.
Voice over IP (VoIP) is voice data transmission service over an Internet protocol (IP) network. Conventionally, voice service has been provided in a circuit switched (CS) domain. In the VoIP, however, the voice service is provided in a packet switched (PS) domain. In CS-based voice services, the voice data is transmitted while connections are maintained in an end-to-end manner. On the other hand, in the VoIP, since the voice data can be transmitted in a connection-less manner, network resources can be very effectively used.
With the development of a wireless communication technique, an amount of user data is rapidly increased. Thus, for effective use of limited network resources, the conventional CS-based services have recently been replaced with PS-based services. The VoIP is being developed in the same vein, and it is expected that all voice services are provided over the VoIP in most of wireless communication systems in the future.
Although the VoIP has an advantage in that network resources can be effectively used, there is also a demerit in that quality of service (QoS) is inferior to that of the CS-based voice services. The QoS is affected by several factors. Representative examples thereof are a delay, a jitter, a high frame error rate (FER), etc. In the initial stage of the development of the VoIP, the QoS was significantly inferior to that of the CS-based service. However, as many researches have been conducted, the VoIP currently used in wired communications ensures almost equivalent QoS as the CS-based voice services.
A real-time transport protocol (RTP) is developed to effectively provide the PS-based voice services. Further, an RTP control protocol (RTCP) is also developed to control the RTP. In the RTP, time stamp information is carried in every packet, and thus a jitter problem can be solved. Further, by reporting loss of an RTP packet through the RTCP, the FER can be reduced through rate control. In addition to the RTP/RTCP, with the development of a session initiation protocol (SIP) and a session description protocol (SDP), virtual connections can be maintained in an end-to-end manner. Therefore, the delay problem can be mostly solved.
At present, in wired communications, satisfactory QoS can be ensured over VoIP. However, in wireless communications, QoS provided over VoIP is far interior to that of the CS-based services. To improve transmission efficiency of the VoIP in a wireless communication environment, an improved header compression scheme (i.e., robust header compression (ROHC)) has been developed and used recently. Still, overall QoS is interior to that of the CS-based voice services.
The RTP and the RTCP are provided by using one stream in wired communications. However, due to different packet characteristics, when the RTP and the RTCP are provided by using one stream in wireless communications, QoS is significantly decreased, which results in one of the biggest problems occurring in the wireless communication system supporting the VoIP. Specifically, the RTP which is real-time user data is insensitive to an error but is sensitive to a delay and a jitter, whereas, the RTCP which is control data is insensitive to the delay and the jitter but is sensitive to the error. Further, in the RTP, since voice data is carried, a small-sized packet is frequently and regularly transmitted. On the other hand, in the RTCP, since control data is carried, a much larger sized packet is transmitted less frequently and less regularly in comparison with the RTP.
Accordingly, there is a need for a method of improving QoS of voice data so that VoIP can be effectively implemented in a wireless communication system.